developing and teaching is’97.2: personal productivity ... · packages effectively to improve...

Volume 1, Article 17June 1999

DEVELOPING AND TEACHING IS’97.2: PERSONALPRODUCTIVITY WITH INFORMATION TECHNOLOGY

Gordon B. DavisJ. David Naumann

Gove AllenCarlson School of Management

University of [email protected]

TUTORIAL

Communications of AIS Volume 1, Article 17 2Developing And Teaching IS’97.2 by G. B. Davis, J.D. Naumann, and G. Allen

DEVELOPING AND TEACHING IS’97.2: PERSONALPRODUCTIVITY WITH INFORMATION TECHNOLOGY

Gordon B. DavisJ. David Naumann

Gove AllenCarlson School of Management

University of [email protected]

ABSTRACT

It is common for university students either to have introductory skills in the

basic desktop software packages or to obtain these skills by self-study modules

or short courses. What students lack is an understanding of how to use desktop

packages effectively to improve their productivity. This is the missing link in the

curriculum. This tutorial describes one solution: the IS’97.2 course on Personal

Productivity with Information Technology.

An appendix presents a sample lesson used with this course.

Keywords: IS curriculum, productivity, desktop computing, knowledge work.

I. INTRODUCTION

The desktop personal computer is fundamental to knowledge work. Over

the years, the software packages have grown in size and scope. For example:

• Microsoft Word version 2 had 311 commands; Word 97 has 1,033

• WordPerfect version 4.2 had 1 megabyte of programs and

supporting materials. Corel WordPerfect 8 includes 100 million


bytes

• PowerPoint 97 has seven pages of user options

• Excel 97 includes 23 different customizable toolbars

When faced with large numbers of features and options in software

packages, users tend to learn a minimum set of commands and never explore

further. Again and again, we find users with significant experience using a

package have not explored options that would be useful to them. Users report to

us that they run only one program at a time. They are confused by hierarchical

file organizations. Students enter college more proficient than ever before. They

can surf the web and use e-mail. Most are comfortable enough with word

processing software to generate a professional-looking resume. By the time they

are admitted to the business school (or not long after), students can solve simple

accounting problems using spreadsheet software and can prepare colorful slide

presentations. These skills are only the foundation of the set that is required to

achieve the potential productivity improvements that information technology

offers knowledge workers. If a student who has been introduced to basic

features and functions wants instruction to help improve productivity in use of

desktop technology, schools usually do not provide a course. Fortunately, such a

course exists. It is defined in the IS curriculum models, and we have had three

years experience in teaching it.

The IS’97 model undergraduate curriculum developed cooperatively by

• ACM (Association for Computing Machinery),

• AIS (Association for Information Systems), and

• AITP (Association for Information Technology Professionals)

contains a new course not previously taught, IS’97.2, Personal Productivity with

Information Technology. The course, defined briefly in the model curriculum, is

described in Appendix A. This tutorial describes the reasoning behind the course

and how it is taught at the University of Minnesota. It includes a suggested

syllabus, course topics, exercises, projects, and teaching methods. The


discussion centers on pitfalls and problems the authors experienced in designing

and delivering of the course, including the source and use of teaching materials.

II. OBJECTIVES

The objective of IS’97.2 is to improve an undergraduate student’s

knowledge work productivity through the effective use of information technology.

Understanding how to apply information technology to achieve improved

productivity in knowledge work is an important element in the general computer

knowledge and skills of all students. It is essential for information system

specialists.

Prerequisite to the IS’97 productivity course is elementary skill in using a

practical “toolkit” of software packages vital to knowledge work. Caution: do not

use the productivity course as a remedial course for the basic toolkit technology;

to do so dilutes its purpose. The toolkit includes software for word processing,

spreadsheets, presentation graphics, database management, Internet access,

and e-mail. The toolkit may be learned in different ways:

• through software skill development course modules,

• in connection with a course making use of the software, or

• by individual self-directed study.

The scope of this instruction generally is limited to basic features and functions.

The IS’97 model curriculum specifies a survey course in concepts,

terminology, and practice for information technology and information systems in

organizations. This survey course is separate from and different in scope and

purpose than the productivity course. It may be taken before or after IS’97.2.

Although these two courses (or course modules) provide basic knowledge

and skill, they are insufficient academic preparation for productive use of

information technology. The IS’97.2 Personal Productivity with Information


Technology course addresses issues and methods not found in either the

prerequisite toolkit or the information systems survey course. It focuses on how

to be more productive in using information resources in individual and

collaborative knowledge work. Knowledge work is defined as:

“human mental work performed to generate useful

information. In doing it, knowledge workers access data, use

knowledge, employ mental models, and apply significant

concentration and attention.” [Davis and Naumann, 1997, p. 6]

When knowledge workers perform individual or collaborative work, their

productivity depends on what they know, their ability to obtain information from a

variety of sources, and their ability to design and carry out analytical and

information processing activities. Information technology improves productivity in

knowledge work.

Organizations are increasingly aware of the importance of knowledge

work productivity. Drucker was prophetic in his 1969 statement:

“To make knowledge work more productive will be the great

management task of this century, just as to make manual work

productive was the great management task of the last century.”

[Drucker, 1969, p. 290]

COURSE OBJECTIVE

The over-all objective for the course is to help students apply information

technology to improve productivity. This goal is achieved by three broad areas of

instruction:


1. Concepts and principles for productivity in knowledge work. Emphasis

is on relating concepts and principles to information technology

features and functions.

2. Analyzing requirements for information technology in knowledge work

and evaluating the productivity effects of alternatives.

3. Developing skill in software functions and features and development

alternatives that provide knowledge work productivity improvement.

These three areas may be modified to suit the knowledge work focus of

students, especially in the problems and exercises employed, and functions and

features studied. For example, accounting students might take more time with

error-control while marketing students might learn more about analytical

functions.

COURSE OUTCOMES

A student who completes the course successfully will have learned to

apply:

• new ways to think about individual and group productivity in knowledge

work,

• new approaches to understanding the value of information technology

in knowledge work,

• new ways to manage the time investment to find and learn new

features in software packages and to achieve payback from the

investment, and

• selected high productivity payoff software package functions and

features.


COURSE ORIENTATION

The course is oriented to packaged software. It assumes that individuals

should always look first for existing solutions such as packages or specialized

applications. If packages or existing applications do not fit, then tailoring or

customizing is recommended. When existing software cannot be tailored to fit a

problem, it may often be extended or replaced with a programmed solution. The

course introduces methods that range from developing simple enhancements of

existing packaged software to complete end user applications. End user

application development is discussed but treated as a less desirable alternative.

This content emphasizes the difference in scope and quality assurance for end

user systems developed for a single user or a small homogeneous group and

high quality, robust applications that can be deployed to multiple unknown users.

The course includes elementary interface development and procedural

programming in a high-level object-based language.

The course provides the future specialist with insight into individual

systems and how they can be built and supported. Also, the course provides a

simplified introduction to principles and procedures for analysis and design. A

“system development in the small” approach is used. This subset of knowledge

and skill supports and enhances learning in subsequent “standard” application

development and programming courses.

III. ASSUMPTIONS ABOUT STUDENT COMPUTER ANDSOFTWARE KNOWLEDGE

We assume that students already know how to use the basic functions

and features of software packages for word processing, spreadsheets,

presentation graphics, e-mail, and the Internet. Some students can also use the

basic capabilities of a database package. The course builds upon this

elementary knowledge and experience. Of course, students will have a wide

range of software package knowledge and skill. Even students with significant


experience do not know the concepts and principles of productivity to guide them

in selecting and using productivity enhancing functions and features. However,

students with prior training and experience with programming, systems

development, and the software package toolkit do not fit the structure and pace

of IS’97.2 and should be provided an alternative path.

IV. APPROACHES TO IMPROVED PRODUCTIVITY

Information technology can be applied to three approaches to improved

knowledge work productivity:

1. Increasing knowledge work efficiency. Many software functions

and features reduce the time and/or energy (attention and

concentration) needed.

2. Increasing effectiveness. Software functions and features support

knowledge workers in extending the scope, depth, and quality of

the work they perform. Many activities are possible with

information technology that are not possible without it, (e.g.,

communication and coordination software, search tools, analytical

functions, error detection/correction functions, and presentation

features)

3. Expansion and/or conservation of the limited time and attention that

knowledge workers can devote to activities. Communications,

coordination, scheduling of work, design of work activities, and

managing motivational factors can conserve or expand the time

and attention that can be applied.

All three approaches can employ economic analysis, although such

analysis is most applicable to efficiency effects. The cost of productivity

improvements is learning time plus lower productivity during the first few uses

(when it may take more time to apply new functionality). The payback is in


subsequent use, through reduced time and energy and through increased

functionality and/or reduced errors.

V. USE OF INFORMATION TECHNOLOGY INPRODUCTIVITY IMPROVEMENT

The course addresses the following principles of productivity

improvement:

• Reuse of structure and procedures

• Efficiency in data access

• Minimizing errors

• Reuse of data • E-mail productivity • Minimizing learning costs

• Efficiency in process activities

• Productivity in collaborative work

• Knowledge work management

REUSE OF STRUCTURE AND PROCEDURES

A major source of productivity is reuse. Most knowledge work activities

and outputs require planning and design as well as execution. The planning of

the work and design of the outputs can often reuse existing patterns and models.

Templates, style sheets, designs, recorded macros, macros, stored queries,

stored reports, custom lists (spreadsheets), and custom expansion of

abbreviations (word processing) are examples of functions that support reuse of

structure and procedures.

REUSE OF DATA

Data used in more than one application should not have to be re-entered.

Examples of functions that support reuse of data are databases, import and

export functions, various copying and pasting options, and linking and

embedding.


EFFICIENCY IN PROCESS ACTIVITIES

Functions and features that support improved efficiency in performing

common activities are outlining and table of contents (word processing), search

and replace, built-in analytical functions, graphing and drawing functions,

macros, and task management functions. Software packages allow options and

preferences. These can be set to match task activities and improve efficiency

and reduce errors. Appropriate use of menus, icons, and keystrokes reduces

access time to functions.

EFFICIENCY IN DATA ACCESS

The time and effort to store and retrieve data can be improved by

directory/file structures, search functions, search planning and stopping rules,

and good database design. Conversion functions among packages and different

file formats reduce time to transfer data.

E-MAIL PRODUCTIVITY

This task is identified separately because of its widespread use.

Productivity can be improved by good folder structure and naming, use of

archiving procedures, filtering mechanisms, directory structures and lists, and

automating functions such as reply and confirmation of receipt.

PRODUCTIVITY IN COLLABORATIVE WORK

Communication and collaboration costs rise rapidly as team size

increases. Use of collaborative work functions can reduce the overhead

associated with teams. Examples are team communications support, scheduling

and coordination functions, shared data access mechanisms, concurrent work

software, collaborative authoring software, and meeting software.


MINIMIZING ERRORS

Errors significantly reduce productivity. Information technology can be

used to reduce errors in knowledge work processes and data access. Examples

are consistent and meaningful naming of files and folders, copying and pasting,

importing and exporting, linking and embedding, automatic entry of phrases and

names based on custom expansion in word processing and custom lists in

spreadsheets, and use of input forms (with validation methods) for entering data.

There are error-detecting functions in databases (such as the examine function

and spreadsheet functions) and practices to reduce and detect errors (such as

the spreadsheet audit function, the use of range names instead of cell

addressing, and freezing panes so column or row headings are visible in large

spreadsheets).

MINIMIZING LEARNING COSTS

A new software package, change, or upgrade has an associated learning

time. If both software features and task process are changed, there is a loss of

productivity in both software use and task activities. When software features are

learned so that they are mainly automatic, attention can be directed primarily to

the task. Process learning time can be reduced by compatible software versions,

by online tutorials, and by online HELP facilities.

KNOWLEDGE WORK MANAGEMENT

Knowledge work is sensitive to motivational factors based on scheduling

of work to provide motivation to complete, planning to reduce rework and lost

opportunities, stopping rules to avoid unproductive expansion of work, search

strategies, and feedback. A combination of organization, management, and

information technology may be applied. Information overload can be reduced by

appropriate data structures, search mechanisms, stopping rules, and data

reduction functions.


VI. EXAMPLE SYLLABUS

Table 1 presents a flow of topics and exercises to achieve the objectives

of IS’97.2. Each topic may take several class and/or lab sessions.

Table 1. Topics and Exercises for IS’97.2

TopicNumber

Lecture/Discussion TopicsProblems on Analysis and SoftwareFunctions, Features, and UserPractices

1 Understanding the nature of knowledgework.Optional: review of core informationtechnology.

Task/activity analysis to analyzerequirements for software packages.

2 How productivity improvement isachieved in knowledge work.

Reuse: templates for wordprocessing, spreadsheets, andpresentations. Style sheets for wordprocessing and presentations.

3 Knowledge work management toimprove productivity.

Software to schedule and trackknowledge work. Software functionsfor switching among tasks.

4 Exploration/evaluation of value oftechnology in collaborative work.

Exploration of collaborative worksoftware. Use of group workfunctions.

5 Analyze data and communicationrequirements. Tools and methods forcommunication productivity.

Using discovery approach torequirements analysis. Applying e-mail tools and methods.

6 Organization and management ofindividual files and directories.

Creating an efficient structure fordirectories and files. File search andindexing tools. Moving data betweenapplications and sharing of data byapplications: copying, importing,linking, embedding.

7 Access and report facilities of databasepackage (for a database that alreadyexists).

Forming and executing queries andreports. Analytical tools for dataanalysis. Importing and exportingto/from spreadsheets.

8 External data search using Internet.Productivity in search procedures andstopping rules.

Applying search methods and tools.Developing and applying stoppingrules to constrain search time andeffort.

9 Selecting options or preferences thatimprove productivity in use of packages.Reuse of sequences of instructions byuse of recorded macros. Reducing timeto access macros and other functions.Functions that reduce effort and reduceerrors. Obtaining assistance from HELPfacilities.

Tailoring and preference options insoftware packages. Recordingmacros for spreadsheets and wordprocessing. Use of menus, icons, andkeystrokes to reduce access time.Reducing effort and errors withautocorrect, fill, and similar functionsExploring efficient use of HELP.

10 Developing a procedure for reuse of asequence of instructions where thereare conditions and alternative

Extend a recorded macro by addinginput box and message boxes Extendone or more recorded macros by


procedures. Program control structures.Adding control structures to recordedmacros.

adding decision logic looping andrepetition until some condition isachieved.

11 Types of user interfaces. Design ofuser interfaces. Software developmentenvironments. Features of a typicaluser interface development tool.

Develop an interface for a simple enduser application and programinstructions for each object a user canselect (using Visual Basic)

12 Evaluating or developing solutions withdatabases.

Develop flat file using databasepackage, create input form, and inputdata. Develop and create databasewith two or more tables and enforceintegrity.

13 Characteristics of robust, error resistantapplications. Methods and functions forerror control and error detection.

Evaluate robustness of an application.Design a spreadsheet that is veryrobust relative to errors. Use ofvalidation rules in database inputforms and database tables.

Optional Technology infrastructure with emphasison infrastructure for individual or smallgroup.

Analyze, specify, and justify apersonal information technologyinfrastructure. Formulating aproductivity-based personal policy onupgrading of software.

Optional Relationship of individual systems toorganizational systems. Organizationalsupport for individual and groupproductivity.

Evaluate cooperation with informationtechnology function.

Project Integration of topics and building of enduser applications to improveproductivity.

Use software packages to createapplication that demonstrates use ofinformation technology to improveknowledge work productivity of anindividual or small group.

VII. TEXTBOOK SUPPORT

The teaching approach combines discussion of concepts and methods

with practical exercises. The concepts and methods provide long-lasting

knowledge; the practical exercises give the student the ability to apply the

knowledge. Details of current technology are skipped or de-emphasized, except

as they are needed to clarify concepts and to support exercises.

The course requires three types of textbook support:

1. concepts and analysis,

2. software functions and features, and

3. exercises.


A book (Davis and Naumann, 1997) was prepared to provide concepts and

analysis, an overview of relevant software functions, and exercises for the

students.

In using the textbook, we found that most students need supplementary

materials:

• A software manual describing how to use the functions and features

emphasized in the course. An ideal solution would be a custom

selection of items from one of the numerous how-to-do software

manuals (especially those that cover some advanced features). An

alternative is to identify HELP pages and examples in the

documentation of the packages chosen.

• Walkthrough examples of use of software functions and features.

These examples may be the basis for class demonstration, but

students need to be able to observe and then repeat them on their

own.

• The instructor must continually emphasize the connection between the

productivity concepts and the functions and features being learned.

Problems and exercises are defined in Table 1. They range from very

simple, low investment, and high payoff functions and features such as templates

and simple recorded macros to more complex exercises such as adding logic to

recorded macros and developing an end user application. Appendix B contains a

comprehensive exercise that illustrates how to do a simple recorded macro for a

spreadsheet and then extend the macro by adding logic. The exercise starts by

defining a repeatable task, records the steps in the process, debugs the macro

until it is right, edits the macro to add loop logic, and stores the resulting macro

for future use.


VIII. ISSUES, PITFALLS, AND PROBLEMS IN COURSEDESIGN AND DELIVERY

The course should be a mix of concepts, analytical approaches, and

software functions. Although concepts and analytical approaches have the

highest potential for long term effect on behavior, they are learned best when

they can be tied to specific software functions and features. Students are

motivated in the course by learning new software functions. They need simple,

concrete demonstrations and hands-on exercises. Principles are learned best

when embedded in exercises; understanding comes from first doing and then

articulating abstract concepts.

To spread out the assignments over the term, the software assignments

must begin very early, even though many concepts have not been presented.

This suggests that very simple, easy to apply functions (such as templates) be

assigned first and more difficult functions be assigned later in the course.

We have taught the course as an elective over three years. The

description for the course indicates it is for those with limited technical

backgrounds. Section sizes have been 20 to 30. Interestingly, students with

technical backgrounds in information systems but without the concepts and skills

taught in the course do better than students with little information systems

background. Students with limited backgrounds often have difficulty in learning

the features taught in the course because they lack the underlying basic skills.

Experience suggests individual projects are useful in achieving course

objectives. The issues include whether students should propose the project or

the instructor should assign. Both approaches have worked for us.

When the course was first offered, we assumed that HELP facilities would

be sufficient for presenting the details of features and functions. This turned out


to not be so. HELP, even with examples, is often insufficient. Although most of

the assignments deal with relatively simple software functions, students need a

walkthrough demonstration. The instructions for features often contain

assumptions or unspecified things that must be done. Completing the first

exercise with new software features can be overly time consuming for students

unless there are detailed instructions for a demonstration and students have

done the demonstration themselves (or at least seen it done).

Database instruction was divided into two parts. The first made the

assumption that the organization provided a database and the student had to

learn to access it and retrieve from it. This part placed the focus on retrieval and

report generation. Issues of database design were avoided. The second part

addressed issues of building a simple end user database. Student assignments

included building a database with at least two tables. The database building

exercise was followed by discussion and exercises related to such topics as error

control, use of data entry forms, and data validation.

Instruction in using software frequently omits any discussion of error

control and practices that reduce error rates. In terms of productivity, errors are

a high-cost type of rework and can have serious consequences if not detected

and removed. Therefore, an important topic addressed the issue of error control.

User practices, functions, and features to reduce input errors and detect errors

were presented and applied in exercises.

End user application development is an optional part of the course Two

types of end user applications are included: database software and a

development language, Visual Basic. Students found the end user development

projects interesting and instructive. However, the time available was not

sufficient to do more than introduce the tools and do simple exercises.


The course worked well when taught in a room in which each student had

a computer and could do simple exercises in real time as a function was

presented. The second best alternative was a live demonstration in class with

students having access to a computer lab after class. The course does not work

well without a live demonstration. This point should be emphasized.

Demonstrations and hands on experience are vital ingredients. With a laboratory

that provides a computer for each student, demonstrations can walk through an

example (with students not using their computers). The demonstration is

followed by an example in which the students repeat the same type of exercise.

This approach is successful because the students reinforce the learning from the

demonstration and also because they help one another.

IX. SUMMARY

The IS’97.2 course, Personal Productivity with Information Technology,

represents an innovative addition to the courses taught by the information

systems faculty, both as part of the set of courses taken by IS majors and also as

a service course. The course extends the elementary skills typically provided in

a software toolkit course or modules. The emphasis is on using information

technology to be more productive as a knowledge worker. The course explains

knowledge work and how productivity is achieved. It presents ways to analyze

requirements for knowledge work and provides practical, hands-on instruction in

software functions and features that typically aid in improving productivity.

This tutorial has described the objectives of the course and assumptions

about the student’s prior computer and software knowledge. Approaches to

improved productivity were summarized to give insight into the nature of the

concepts for the course. Ways in which information technology can be used to

achieve productivity improvements were presented along with examples of

processes and software functions and features to be taught in the course. A

sample syllabus of topics and exercises illustrate a possible structure for the


course. The issue of textbook support was explored. Insights obtained from

offering the course were noted.

EDITOR’S NOTE: This article was received on April 1,1999. It was with the authorsapproximately 2 months for revision. It was published on June 30, 1999. The article is based on atutorial presentation delivered by the authors at the Fourth Americas Conference on InformationSystems, August 14-16, 1998, Baltimore, Maryland

REFERENCES

Davis, G. B., J. T. Gorgone, J. D. Couger, D. L. Feinstein, H. E.

Longenecker Jr. (1997), IS’97 Model Curriculum and Guidelines for

Undergraduate Degree Programs in Information Systems, Jointly Sponsored by

the Association for Computing Machinery, Association for Information Systems,

and Association of Information Technology Professionals.

Davis, G. B., and J. D. Naumann (1997), Personal Productivity with

Information Technology, New York: McGraw-Hill.

Drucker, P. F. (1969), The Age of Discontinuity, New York: Harper & Row.

APPENDIX A

IS’97 AVAILABILITY ANDIS’97.2 COURSE DESCRIPTION

AVAILABILITY

The IS’97 curriculum report (Davis et al., 1997), is available through the

three participating organizations, in printed form from two of the organizations

and accessible on the web in both pdf format and html from the third.

• Included in the ACM journal, Database, Vol. 28, No. 1, Winter 1997.


• Available as a report from the Association for Information Technology

Professionals (AITP), (formerly DPMA). See their website for

instructions on ordering: http://www.aitp.org.

• Accessible electronically (AIS) website: http://www.aisnet.org/. Select

Resources and then choose either of the following:

a. The link to the portable document format (pdf) and download using

Adobe Acrobat. The document is about 100 pages.

b. The link to the htm version and read using the web browser.

COURSE DESCRIPTION

IS’97.2 — Personal Productivity with IS Technology (Prerequisite: IS’97.PO)

Scope: This course enables students to improve their skills as knowledge

workers through effective and efficient use of packaged software. It

covers both individual and group work. The emphasis is on

productivity concepts and how to achieve them through small

systems.

Topics: End user systems versus organization systems; analysis of

knowledge work and its requirements; knowledge work productivity

concepts; software functionality to support personal and group

productivity; organization and management of software and data;

accessing organization data; accessing external data; selecting a

computer solution; developing a macro program by doing; designing

and implementing a user interface; developing a solution using

database software; refining and extending individual and group

information management activities.


APPENDIX B

A STUDENT EXERCISE ILLUSTRATING THREEPRODUCTIVITY PRINCIPLES AND FEATURES

(SENSITIVITY ANALYSIS USING EXCEL 97)

This exercise covers three features that could be presented separately.

Presenting them in the same exercise illustrates how various features can be

used together to build a complete solution.

A sensitivity analysis consists of set of formulas that rely on one or more

input values and then generate one or more results. For example, a person

might create a worksheet to predict the contribution margin of a proposed new

product. The model uses direct cost, price, and quantity as inputs and the

contribution margin as the output (Figure 1).

Figure 1. Simple Sensitivity Analysis

This worksheet allows a knowledge worker to quickly determine the

expected contribution margin for a new product if the direct costs are 40 instead

of 35, or if the quantity sold is only 500 instead of 5,000, or if the price needed to

sell 5,000 is 85 instead of 100. This model works well for best-case and worst-

case scenarios; however, the best and worst are not what we expect to happen.

They do not represent our best guess of what contribution margin the new

product will provide, nor do they indicate a level of confidence about the best

guess.

Suppose our experience with price and cost estimates has been that 95

percent of the time the estimate is within 10 percent of the estimated price and


that the deviation from the estimates are normally distributed. In this situation,

we expect the cost to be $35; however, we are much more comfortable to assert

that the price will be between $31.50 and $38.50. Likewise, we expect the price

needed to sell 5,000 units will be between $90 and 110. How do we express our

best guess at the contribution margin this product will generate?

One way to do this is by simulation. We can describe the cost and price

parameters as functions that get their values from a particular distribution. We

can recalculate the worksheet several times to see the different values generated

by the formula for the contribution margin. Using Excel’s norminv function, we

can select a value from a normal distribution based on a relative position on the

specified normal curve. If we combine the norminv function with the rand

function, we can let Excel decide which part of the normal curve to use in

selecting the cost and price for a particular observation of the projected

contribution margin. Figure 2 shows the formulas used to randomly generate the

cost and price for the simulation.

Figure 2. Sensitivity Analysis Configured for Simulation

The next step is to recalculate the worksheet many times and to record

the different values observed for the contribution margin. In order to do this, we

need spreadsheet cells to store the observed data. In this example, column D is

used for the observation data. After each recalculation (F9 forces a

recalculation), we need to store the observed contribution margin. To do this in

this example, the user must select the cell (B4) containing the result and select


“Copy” from the “Edit” menu. Move to cell D1 and select “Paste Special” from the

“Edit” Menu. This will invoke the dialog box shown in Figure 3.

Figure 3. “Paste Special” Dialog Box

The user must select “Values” as highlighted in Figure 3, then click the

“OK” button. This will paste the value generated by a particular combination of

possible values for both price and cost. After the user has done this process 10

times the worksheet will have 10 values as shown in Figure 4.

Figure 4. Simulation Results from 10 Observations


Each of these 10 observations represents a possible contribution margin

for the proposed new product. Although the observations presented will not

exactly match the observations of another user (because of the use of the

random function), they should be similar. A quick review of these observations

indicates that the highest value observed is over $459,000 and the lowest is less

than $178,000. Although these values give us some idea of a range, a better

way to describe this data is with its mean and confidence interval. We can

calculate the mean by using the “average” function. Since a 95 percent

confidence interval is 1.96 standard deviations on each side of the mean, we will

calculate an upper and a lower boundary using the “stddev” function. These

functions are shown in Figure 5.

Figure 5. Formulas for Mean and Confidence Interval

The formulas shown in Figure 5 use (D:D) to indicate the entire column.

As a result, as more observation are generated, they will automatically be taken

into account in the mean and confidence interval calculations. Figure 6 shows

the values calculated for mean and confidence interval for 10 observations of the

contribution margin.


Figure 6. Values for Mean and Confidence Interval

With 10 potential outcomes, the expected contribution margin is about

$335,000. The confidence interval is quite wide, being greater than $300,000.

Since the standard deviation becomes smaller as the number of observations

increases, the estimate is improved by making many more observations.

Although we could add observations by continuing the process of recalculating

the worksheet and then copying and pasting the values into the data column, we

can automate that process, in part to speed up the process of recording various

observations, but also to reduce the possibility of operator error.

In expanding the basic spreadsheet sensitivity analysis, three features of

spreadsheets will be employed. They illustrate they way productivity may be

improved both by increasing efficiency and by improving the usefulness of the

solution. The three features to be implemented are as follows:

Feature Productivity principleRecord Macro to select and storeobservations

Reuse of a procedure

Assign a keyboard shortcut to run macro Efficiency (using keystroke preferences)

Add a loop to a recorded macro to morefully automate a procedure

Efficiency (automate loop)Effectiveness (make a more general, moreuseful application)


A recorded macro saves a set of keystrokes and/or menu commands that

can later be invoked to perform a particular task. To record this macro, use the

following steps:

1. Select cell A1. (This is just to make sure we are starting at the same place)

2. Begin recording a macro by choosing [Tools | Macro | Record New Macro]

from the menu bar as shown in Figure 7.

Figure 7. Recording a Macro

3. In the “Record Macro” dialog box, enter a name for the macro name and

capital “O” as the short cut key, then click “OK” as shown in Figure 8. This step

allows the macro to be executed either by selecting the name from the list of

macros (by choosing [Tools | Macro | Macros] form the menu bar) or by pressing


ctrl+shift+O from the keyboard. (Assigning this macro to a keyboard shortcut is

an example of increasing productivity by setting efficiency preferences.)

Figure 8. Record Macro Dialog Box

4. All operations from this point until stop will be recorded into the macro.

To make room in a list for each observation starting at D1, select cell

D1 then choose [Insert | Cell] from the menu bar. The Insert dialog box

is invoked, as shown in Figure 9. Select “Shift Cells Down” and then

click “OK.”

Figure 9. Insert dialog box—Making Room for More Data

5. Press F9 on the keyboard to force the worksheet to recalculate

6. Copy the new observation of the contribution margin (from cell B6)

7. Select Cell D1 and paste the value of the newly copied contribution margin

observation (Edit | Paste Special | Values)


8. The macro is now complete. Click the stop button (the button on the left in

Figure 10) on the record macro toolbar. This tool bar appears while recording a

macro.

Figure 10: Ending the Macro

The user invokes (plays) the macro by pressing ctrl+shift+O to generate

an observation and record it. By pressing the keys to play the macro 89 more

times (for a total of 100 observations), the width of the confidence interval for the

data is reduced from over $300,000 to under $220,000, as shown in Figure 11.

Figure 11. Confidence Interval after 100 Observations

If the user wishes to generate a large number of observations (such as

1000) it makes sense to modify the macro further. The modification will reduce

the time needed to run the analysis and reduce possible errors in using the

procedure; it also makes a more complete user-written application.

The modification is to edit the Visual Basic code generated by recording

the macro. The user to selects [Tools | Macro | Macros] from the menu bar. This

invokes the Macro dialog box as seen in Figure 12.


Figure 12. Macro Dialog Box

From the macro dialog box, the user selects the macro named

“Observation” and clicks the “Edit” button. The Visual Basic code for the macro

appears in an editor as shown in Figure 13.

Figure 13. Visual Basic Editor and Code for Observation Macro


Using the Visual Basic editor, the user adds a loop around the

recorded statements to make them repeat the desired number of times. Figure

14 shows a “For…Next loop” added to the code to make it iterate 1,000 times.

Figure 14. For...Next Loop

When the macro with the loop is run, it will generate 1,000 new

observations. In other words, editing the Visual Basic code recorded by the

macro recorder, we have changed the set of instructions from a single recorded

macro to the beginnings of an end user application.

ABOUT THE AUTHORS

Gordon B. Davis holds an endowed chair as Honeywell Professor of

Management Information Systems in the Carlson School of Management at the

University of Minnesota. He is a pioneer in the field and has been active in the

major initiatives to establish the academic discipline. His current research

interests are conceptual foundations for the field of information systems,

management of the IS function, and personal productivity with information

systems.


J. David Naumann is associate professor of Management Information

Systems in the Carlson School of Management at the University of Minnesota.

He is an alumnus of the University of Minnesota Professor Naumann's research

focuses on the process of information systems development, and he has

published articles in the leading MIS journals. He co-authored the textbook

Personal Productivity with Information Technology (McGraw-Hill) with Gordon B.

Davis. He is active nationally in MIS curriculum development, as a member of

the IS Curriculum '97 Task Force sponsored by the three leading MIS

professional organizations: ACM, DPMA and AITP. He designed and operates

the IS Faculty Directory, a World Wide Web international database of 4,000 IS

researchers. He teaches telecommunications at both the MBA and

undergraduate levels. He also teaches applications development and systems

analysis and design courses.

Gove N. Allen is a doctoral student in the Information and Decision

Sciences Department, Carlson School of Management, University of Minnesota.

He received a BS in Accounting and an MAcc with an emphasis in Information

Systems from Brigham Young University in 1994. His primary teaching

responsibilities and research interests are in the areas of information system

development, logical database design, management of intellectual property, and

user interface design. From 1993 to 1997, he operated a training and consulting

firm specializing in relational database design and implementation and object-

oriented client/server development environments, working with such companies

as Sony, AT&T, Sprint, MCI, NASA, Kennedy Space Center, The Money Store,

American Express, and Lincoln Life.

Copyright ©1999, by the Association for Information Systems. Permission to make digitalor hard copies of all or part of this work for personal or classroom use is granted without feeprovided that copies are not made or distributed for profit or commercial advantage and thatcopies bear this notice and full citation on the first page. Copyright for components of this workowned by others than the Association for Information Systems must be honored. Abstracting withcredit is permitted. To copy otherwise, to republish, to post on servers, or to redistribute to listsrequires prior specific permission and/or fee. Request permission to publish from: AISAdministrative Office, P.O. Box 2712 Atlanta, GA, 30301-2712 Attn: Reprints or via e-mail [email protected]

mailto:[email protected]


EDITORPaul Gray

Claremont Graduate University

AIS SENIOR EDITORIAL BOARDHenry C. Lucas, Jr.Editor-in-ChiefNew York University

Paul GrayEditor, CAISClaremont Graduate University

Phillip Ein-DorEditor, JAISTel-Aviv University

Edward A. StohrEditor-at-LargeNew York University

Blake IvesEditor, Electronic PublicationsLouisiana State University

Reagan RamsowerEditor, ISWorld NetBaylor University

CAIS ADVISORY BOARDGordon DavisUniversity of Minnesota

Richard MasonSouthern Methodist University

Jay NunamakerUniversity of Arizona

Henk SolDelft University

Ralph SpragueUniversity of Hawaii

CAIS EDITORIAL BOARDSteve AlterUniversity of SanFrancisco

Barbara BasheinCalifornia StateUniversity

Tung BuiUniversity of Hawaii

Christer CarlssonAbo Academy, Finland

H. Michael ChungCalifornia State University

Omar El SawyUniversity of SouthernCalifornia

Jane FedorowiczBentley College

Brent GallupeQueens University, Canada

Sy GoodmanUniversity of Arizona

Chris HollandManchester BusinessSchool, UK

Jaak JurisonFordham University

George KasperVirginia CommonwealthUniversity

Jerry LuftmanStevens Institute ofTechnology

Munir MandviwallaTemple University

M.Lynne MarkusClaremont GraduateUniversity

Don McCubbreyUniversity of Denver

Michael MyersUniversity of Auckland,New Zealand

Seev NeumannTel Aviv University,Israel

Hung Kook ParkSangmyung University,Korea

Dan PowerUniversity of Northern Iowa

Maung SeinAgder College, Norway

Margaret TanNational University ofSingapore, Singapore

Robert E. UmbaughCarlisle ConsultingGroup

Doug VogelCity University of HongKong, China

Hugh WatsonUniversity of Georgia

Dick WelkeGeorgia StateUniversity

Rolf WigandSyracuse University

Phil YettonUniversity of New SouthWales, Australia

ADMINISTRATIVE PERSONNELEph McLeanAIS, Executive DirectorGeorgia State University

Colleen BauderSubscriptions ManagerGeorgia State University

Reagan RamsowerPublisher, CAISBaylor University

developing and teaching is’97.2: personal productivity ... · packages effectively to improve...

Documents